summaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2009-12-04Use signalfd for signal handlingDenis Kenzior1-32/+58
2009-12-04Make sure to list calls in HFP voicecall driverDenis Kenzior1-0/+18
2009-12-04Fix release clcc_source when voicecall is removedZhenhua Zhang1-0/+3
Release the timer to avoid invoking the callback function after voicecall driver is removed.
2009-12-04Add hfp_send_dtmf to send tones for hfpmodemZhenhua Zhang1-1/+40
2009-12-04Optimize: call list signal emissionDenis Kenzior1-9/+0
2009-12-04Refactor: Run CLCC when initializingDenis Kenzior1-56/+9
When voicecall atom is initialized, we should run a CLCC query to populate our call list. This also notifies the core of the new calls
2009-12-04Update atmodem voicecall driver to new world orderDenis Kenzior1-9/+7
2009-12-04Fix: StyleDenis Kenzior1-2/+1
2009-12-04Update HFP voicecall driver to new world orderDenis Kenzior1-10/+7
2009-12-04Remove: at_util callid APIs are no longer necessaryDenis Kenzior2-22/+0
2009-12-04Update voicecall atom to use new callid apiDenis Kenzior1-9/+7
2009-12-04Get rid of list_calls in hfp driverDenis Kenzior1-1/+0
2009-12-04Get rid of list_calls in calypso driverDenis Kenzior1-1/+0
2009-12-04Get rid of list_calls in isi driverDenis Kenzior1-16/+0
2009-12-04Add utility to get the next free call idDenis Kenzior2-0/+12
2009-12-04Refactor: call id allocation / releaseDenis Kenzior2-5/+10
2009-12-04Refactor: Remove list_calls from voicecall driverDenis Kenzior1-2/+0
2009-12-03Mention Martin's contributionsMarcel Holtmann1-0/+1
2009-12-03Allow builds to install the test scripts for debuggingMartin Xu3-2/+12
2009-12-03Fix: Exit right awayDenis Kenzior1-0/+3
If no plugin needs to run the eventloop to shutdown cleanly, exit right away.
2009-12-03Add unit test for max message size encodingDenis Kenzior1-0/+61
2009-12-02Fix: Be a bit more signal-handler safeDenis Kenzior1-5/+12
2009-12-02Fix: Paste errorDenis Kenzior1-2/+0
2009-12-02When SIGTERM is received, run the event loopDenis Kenzior3-13/+71
This feature will continue running the event loop for another 10 seconds or so or until all the modems have been shut down. This will allow the modem drivers to properly shutdown their respective devices.
2009-12-02Fix: Convert PalmPre plugin to new world orderDenis Kenzior1-0/+4
2009-12-02Fix: Convert Novatel plugin to new world orderDenis Kenzior1-0/+3
2009-12-02Fix: Convert MBM plugin to new world orderDenis Kenzior1-0/+3
2009-12-02Fix: Convert Huawei plugin to new world orderDenis Kenzior1-0/+3
2009-12-02Fix: Convert HSO plugin to new world orderDenis Kenzior1-4/+8
2009-12-02Fix: Convert G1 plugin to new world orderDenis Kenzior1-0/+9
2009-12-02Refactor: Set powered to false if disable timedoutDenis Kenzior1-1/+12
2009-12-02Add g_at_chat_unregister_all functionDenis Kenzior2-0/+29
2009-12-02Fix: unregister modem when unregistering driverDenis Kenzior1-0/+12
2009-12-02Refactor: modem de-registrationDenis Kenzior1-9/+3
2009-12-02Fix: Remove all atoms when attempting to power offDenis Kenzior1-9/+9
The current behavior is to remove atoms when the modem has powered off successfully. The modem driver must now take special precautions to cancel all pending actions of the atom drivers, since they would have been destroyed.
2009-12-02Add: g_at_chat_cancel_all functionDenis Kenzior2-0/+28
This will cancel all commands currently queued.
2009-12-02Fix: Sometimes it is possible to cancel first cmdDenis Kenzior1-1/+2
Commands on the head of the queue can be canceled if none of its bytes have actually been written to the modem. This is tracked by cmd_bytes_written variable.
2009-12-02Fix: Don't allow cancelation of command id 0Denis Kenzior1-0/+4
2009-12-02Release 0.12Marcel Holtmann2-1/+12
2009-12-02Fix: Store refs as 16 bits, not 8Denis Kenzior1-1/+1
Refs are handled as 16 bits properly everywhere, yet are stored as 8 bits in the assembly.
2009-12-01Fix: Don't forget to cleanup the atom watchDenis Kenzior1-0/+7
2009-12-01Fix: Make sure to never overflow DNS variableDenis Kenzior1-2/+5
2009-12-01Fix: Poll ENAP on older mbm hardwareDenis Kenzior1-125/+186
2009-12-01Style: Fix whitespaceDenis Kenzior1-1/+1
2009-12-01Fix: Fix parsing in MBMDenis Kenzior1-0/+3
2009-12-01Add experimental support for E2IPCFGDenis Kenzior1-20/+163
2009-12-01Fix: Don't decode error unnecessarilyDenis Kenzior1-2/+2
2009-12-01Fix: Add username/password support to mbm contextDenis Kenzior1-1/+7
2009-12-01Fix: Make sure we don't overflow the bufferDenis Kenzior2-2/+8
2009-12-01Fix: Force calypso to return to command mode ASAPDenis Kenzior1-2/+2
Calypso firmware is just bizarre. It seems that if anything is sent before CPI=3 to the DLC handling voice commands, the ATD will not return to command mode immediately, but instead wait until connected / failed condition occurs. If CPI=3 is sent first, then calypso will go to command mode immediately after ATD is issued and report CPI indications as usual.